sort-union-types

Enforce sorted TypeScript union types.

Adhering to this rule ensures that union types are consistently sorted, resulting in cleaner and more maintainable code. By promoting a standardized ordering of union types, this rule makes it easier for developers to navigate and understand the structure of type unions within the codebase.

Consistently sorted union types enhance the overall clarity and organization of your code.

Important

If you use the sort-type-constituents rule from the @typescript-eslint/eslint-plugin plugin, it is highly recommended to disable it to avoid conflicts.

Try it out

Options

This rule accepts an options object with the following properties:

type

_{default: 'alphabetical'}

Specifies the sorting method.

'alphabetical' — Sort items alphabetically (e.g., “a” < “b” < “c”) using localeCompare.
'natural' — Sort items in a natural order (e.g., “item2” < “item10”).
'line-length' — Sort items by the length of the code line (shorter lines first).

order

_{default: 'asc'}

Determines whether the sorted items should be in ascending or descending order.

'asc' — Sort items in ascending order (A to Z, 1 to 9).
'desc' — Sort items in descending order (Z to A, 9 to 1).

ignoreCase

_{default: true}

Controls whether sorting should be case-sensitive or not.

true — Ignore case when sorting alphabetically or naturally (e.g., “A” and “a” are the same).
false — Consider case when sorting (e.g., “a” comes before “A”).

specialCharacters

_{default: keep}

Controls whether special characters should be trimmed, removed or kept before sorting.

'keep' — Keep special characters when sorting (e.g., “_a” comes before “a”).
'trim' — Trim special characters when sorting alphabetically or naturally (e.g., “_a” and “a” are the same).
'remove' — Remove special characters when sorting (e.g., “/a/b” and “ab” are the same).

locales

_{default: 'en-US'}

Specifies the sorting locales. See String.prototype.localeCompare() - locales.

string — A BCP 47 language tag (e.g. 'en', 'en-US', 'zh-CN').
string[] — An array of BCP 47 language tags.

partitionByComment

_{default: false}

Allows you to use comments to separate the members of union types into logical groups. This can help in organizing and maintaining large union types by creating partitions within the enum based on comments.

true — All comments will be treated as delimiters, creating partitions.
false — Comments will not be used as delimiters.
string — A regexp pattern to specify which comments should act as delimiters.
string[] — A list of regexp patterns to specify which comments should act as delimiters.

partitionByNewLine

_{default: false}

When true, the rule will not sort the members of an union type if there is an empty line between them. This can be useful for keeping logically separated groups of members in their defined order.

type CarBrand =
  // Group 1
  Fiat |
  Honda |

  // Group 2
  Ferrari |

  // Group 3
  Chevrolet |
  Ford

Each group of union types (separated by empty lines) is treated independently, and the order within each group is preserved.

newlinesBetween

_{default: 'ignore'}

Specifies how new lines should be handled between union type groups.

ignore — Do not report errors related to new lines between union type groups.
always — Enforce one new line between each group, and forbid new lines inside a group.
never — No new lines are allowed in union types.

This options is only applicable when partitionByNewLine is false.

groups

_{type: Array<string | string[]>}_{default: []}

Allows you to specify a list of union type groups for sorting. Groups help organize types into categories, making your type definitions more readable and maintainable.

Predefined groups:

'conditional’ — Conditional types.
'function’ — Function types.
'import’ — Imported types.
'intersection’ — Intersection types.
'keyword’ — Keyword types.
'literal’ — Literal types.
'named’ — Named types.
'object’ — Object types.
'operator’ — Operator types.
'tuple’ — Tuple types.
'union’ — Union types.
'nullish’ — Nullish types (null or undefined).
'unknown’ — Types that don’t fit into any group entered by the user.

If the unknown group is not specified in the groups option, it will automatically be added to the end of the list.

Each union type will be assigned a single group specified in the groups option (or the unknown group if no match is found). The order of items in the groups option determines how groups are ordered.

Within a given group, members will be sorted according to the type, order, ignoreCase, etc. options.

Individual groups can be combined together by placing them in an array. The order of groups in that array does not matter. All members of the groups in the array will be sorted together as if they were part of a single group.

Example 1

Using all predefined groups:

type Example =
  // 'conditional' — Conditional types.
  | (A extends B ? C : D)
  // 'function' — Function types.
  | ((arg: T) => U)
  // 'import' — Imported types.
  | import('module').Type
  // 'intersection' — Intersection types.
  | (A & B)
  // 'keyword' — Keyword types.
  | any
  // 'literal' — Literal types.
  | 'literal'
  | 42
  // 'named' — Named types.
  | SomeType
  | AnotherType
  // 'object' — Object types.
  | { a: string; b: number; }
  // 'operator' — Operator types.
  | keyof T
  // 'tuple' — Tuple types.
  | [string, number]
  // 'union' — Union types.
  | (A | B)
  // 'nullish' — Nullish types.
  | null
  | undefined;

groups option configuration:

{
  groups: [
    'conditional',
    'function',
    'import',
    'intersection',
    'keyword',
    'literal',
    'named',
    'object',
    'operator',
    'tuple',
    'union',
    'nullish',
  ]
}

Example 2

Combine and sort intersection and union groups together:

type Example =
  | AnotherType // 'named'
  | SomeType    // 'named'
  | (A & B)     // 'intersection'
  | (A | B)     // 'union'
  | (C & D)     // 'intersection'
  | (C | D)     // 'union'
  | keyof T;    // 'unknown'